Security audit

A股多条件筛选

Security checks for vulnerabilities and agentic risk

Overview

The skill is a finance-analysis guide with no executable code, but it asks for broad command execution, API-key use, external data access, and result exports without enough scoping or user-control detail.

Install only if you are comfortable letting the agent run finance-data commands and contact external data APIs. Use scoped API keys in environment variables, avoid account-trading credentials, require confirmation before command execution or exports, and choose explicit output paths that do not overwrite important or sensitive files.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Intent-Code Divergence

Medium

Confidence: 93% confidence
Finding: The documentation states the skill can be used as pure SKILL.md with no code execution required, yet the manifest declares the exec tool. This mismatch can mislead users and agents into granting or trusting execution capability they would otherwise avoid, increasing the chance of unintended command execution in a finance-oriented workflow.

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The skill claims risky code was removed and security was enhanced, but it still advertises executable capability. Security-assurance language combined with retained execution privileges can create false trust and reduce scrutiny, making social-engineering style abuse more plausible if later prompts or linked behavior trigger shell commands.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The skill describes exporting results to files but does not clearly warn that this writes to local disk, may overwrite files, or may place sensitive financial data in insecure locations. In an agent environment, silent file creation/export can expose data or cause unintended modification of the user's workspace.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill instructs users to configure API keys and use external financial data APIs without clearly disclosing outbound network access, third-party data sharing, or secret-handling expectations. This can lead to unintentional transmission of user queries, credentials, or financial analysis context to external services.

Static analysis

No suspicious patterns detected.